[alma-config] topography mask

[Bryan Butler]

as to first vs. second derivatives, for the question of where
you can put a particular single antenna, the first derivative
seems to me to be the proper quantity.  for location of the
entire configuration (other than perhaps the compact config),
the tilt should be relatively unimportant.

on the question of making a "cost mask" vs. a "gradient mask"
vs. a "binary mask", my feeling is that the cost mask is preferable,
but i suspect that it will be impossible to achieve within a
reasonable amount of time.  we will waste too much time arguing
about the cost model, rather than spending time on more important
issues, IMHO.  the "gradient mask" is easy to do, but i hesitate
to do it, because i would rather have us all use exactly the same
mask.  let's choose the limit and make the mask.  i am currently
making masks for 2 limits of local slope/gradient (including a
shadowing constraint, science preserve boundary constraint, and
pipeline proximity constraint), and will make them available
later today (probably - depends on how much time i spend in the
SSR meeting) for your enjoyment, at least as a first cut.

	-bryan